Dissertation > Excellent graduate degree dissertation topics show

Research on Binocular Vision Stereo Matching Algorithm

Author: XuYuan
Tutor: CaoMaoYong
School: Shandong University of Science and Technology
Course: Control Theory and Control Engineering
Keywords: Binocular stereo vision Calibration Feature extracting Stereo matching
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 439
Quote: 0
Read: Download Dissertation

Abstract


Stereo Vision is one of the most famous researches in Computer Vision Area, and it has widely application requirements in many fields, such as virtual reality, machine vision, multimedia education, digital entertainment, appearance design of industrial products, sculpture and architecture etc. However, because of its numerous information and difficulty in extracting features and inaccurate feature descriptions and other factors lead to inaccurate image matching, thus affecting the stereo vision in practical application. Therefore, the study of robust stereo vision matching algorithm will have important theoretical and practical significance on three-dimensional scene reconstruction.After the reading of literature at home and abroad, I has known the status of binocular stereo vision. In Framework of Marr theory, research binocular stereo vision system calibration, image rectification, stereo matching,3D reconstruction, etc.Details are as follows:First of all, for the pinhole camera model, the traditional three-dimensional calibration, self-calibration, Zhang Zhengyou plane calibration, flexible, simple, stable calibration achieved Zhang’s camera calibration of the system. Calibrate the relative position between two cameras by triangulation algorithm. Secondly, the use of the camera calibration matrix obtained calibration images, and grayscale images after correction converter, low pass filtering, image enhancement and a series of preprocessing, get little noise, high-quality images with strong contrast. On this basis, using Harris corner detection and Canny edge detection obtained useful image information, and there corners and edge points on the image to be used as matching point. Then, after analyzing the fundamental matrix of the 8 points algorithm and 7 points algorithm, choose an effective method to obtain the fundamental matrix of the system. According to the characteristics of stereo matching as well as basic knowledge of geometry, using normalized cross-correlation function to accomplish image matching. In order to get accurate matching results using the fundamental matrix and the relationship between corresponding points match and the principle of two-way matching constraints, to optimize the results of the initial match and eventually obtain the global optimum result.Finally, matching the corresponding points obtained earlier, the basic matrix, intrinsic matrix three-dimension reconstruction, three-dimensional information obtained with the space. Stereo vision through the programming platform built to verify the validity of the whole algorithm.

Related Dissertations

  1. Application of high pressure common rail diesel engine technology in the development of,TK42
  2. 14-bit 100 MB samples / sec pipelined ADC Research and Design,TN792
  3. HD image license plate location based vehicle tracking and violation detection,TP391.41
  4. Stereoscopic three-dimensional reconstruction -related technology research and implementation,TP391.41
  5. Light-section method vision measurement system, the key technology research and implementation,TP391.41
  6. Camera calibration Related Issues,TP391.41
  7. Compressed air flow motion detection circuit breaker,TM561.1
  8. Based on the visual system of unmanned autonomous helicopter landing,V279
  9. Long focal length collimator vacuum aberration measuring and calibration methods,TH74
  10. Growth in high-precision analog measurement circuit accuracy Method,TN710
  11. Hyperspectral imager radiometric calibration laboratories,TH744.1
  12. Hull section robotic welding path planning and offline programming,TP242
  13. Based on Machine Vision Industrial robots crawling technology research,TP242.2
  14. Array AOI system error correction algorithm design and implementation,TP274
  15. The DBF array of amplitude and phase error calibration robustness of the algorithm,TN820
  16. Raster-based three-dimensional projection of dental measurement system,TP391.41
  17. A Dome projection system design and implementation,TP391.41
  18. Media terminal touch screen design and implementation,TP334.3
  19. A new dual- LTE-A and robustness of the design,TN929.5
  20. VHF UCA Research and implementation of active calibration scheme,TN911.7
  21. The Design and Analysis of Electronic Control System and Test Rig for Single-cylinder Motorcycle Engine,U483

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile